This page gives a breakdown of the years with the Port Elizabeth Technikon, briefly describes the Electronic Engineering Department, subjects lectured plus the staff and students I was fortunate to have met.
I had supplied custom designed hardware for Hybrid Energy research around 1990, and shortly afterwards was a promoter for a group of students. I lectured part-time from 1992, full-time between 2003 and 2006, and part-time in 2007.
The students were involved in vocational training at a level between technical colleges and university degrees. The qualification was a National Diploma in Electronics, which was several semester blocks or two years fulltime. Once qualified, the students did a year of “in service” training. The university B.Sc. engineering degrees were four years full-time, as South African schooling system did not have the UK equivalent of A-levels. The first year of a B.Sc was essentially A-levels (making it a breeze for students from Zimbabwe in the 1980s). Around 1992, the Master's Diploma was introduced at Technikons where the Diploma graduate students did a project, wrote up a thesis and presented their work.
After the 1994 elections, a priority was to improve technical skills—South Africa was degree top-heavy with fewer apprentices due to miserable salaries and long hours. The Technikons were allowed to award B.Tech degrees to pave the way for diploma students to progress to degrees. First introducing the B.Tech and M.Tech, then a little controversial, the D.Tech which was called a Doctorate with the title of Doctor, even though the universities in South Africa did not recognise the B.Tech or M.Tech for entry into university at the same level or year.
Instead, universities essentially made the students redo the core subjects, which was equivalent to redoing the engineering degree. Over a fifteen year period and three changes in ministers of education, some universities were forced to merge with the Technikons. The more prestigious or better funded universities refused, but as Port Elizabeth University and Technikon were more reliant on government subsidies, they had little choice.
Fortunately, there was no engineering faculty at the university so there was little resistance (from an engineering standpoint), and the Technikon was able to maintain their diplomas, B.Tech and M.Tech degrees. The combined institution, renamed the Nelson Mandela Metropolitan University was allowed to grant a B.Eng degree. This was completely separate from the diploma, B.Tech or M.Tech streams without a migration path between the two (much like the previous resistance of universities to accept credits for their batchelors degrees). This was a welcome addition of a Mechatronics degree that would be accredited by the Engineering Council and fill a need for local industry plus students did not have to travel 750km to the nearest university offering an engineering degree (B.Sc. Eng).
The B.Tech subjects were not offered each year at PE Tech, as the students mostly attended evening classes which ran for a semester with three contact hours per week per subject (a different subject each evening for four nights per week). The B.Tech was presented as a year full-time course at some of the other Technikons, but most of the students in the Port Elizabeth area were working in industry who allowed them to leave earlier by starting work earlier. The part-time B.Tech took two years, with a fairly large project done in the last year. The project was mostly work related, benefitting both parties. (Confidential projects were examined by the hosting companies with some Technikon staff involvement, but the theses were not placed on the University website—VW, catalytic converters, and process specific projects as Port Elizabeth was the automotive hub in South Africa).
Several Technikon graduates for whom I had been a promoter, co-promoter, lecturer or external examiner, went overseas to cross the diploma to degree boundary as mature students, and obtained PhD's. Well done!
The Technikon invited industrial partners to attend curriculum meetings, particularly those who placed students in recognised training. My involvement was due to the equipment designed for students and lecturers, plus being a promoter/ examiner for the Masters Diploma course.
I supervised M.Dip students in 1990 for Andries van der Linder in the School of Electrical Engineering, developing hardware for the Hybrid Energy project sponsored by Eskom; the South African electricity supplier. I helped to establish a toolchain, development environment, Makefiles and provided sample code for the 68000 based data logger I had developed earlier. I wrote the firmware, real-time kernel, several interfaces and tested the I/O. Each M.Dip student was given the code and a section to develop further.
Several “Masters Diplomas” graduated, establishing the Technikon’s Hybrid Energy Research capability. The award recognised the voluntary work, (the hardware was paid for).
The specification of the 68000 based data logger for recording weather conditions (temperature, wind speed), wind generator control and solar panel research can be found here
Two researchers wanted a high speed data logger that they could program for monitoring power lines. I quoted on a system, then took four months to deliver (schematic capture, PCB layout, assemble, test, reverse engineer the B004 Inmos link board and demonstrate loop sampling a sine wave at 3 MHz simultaneously on four channels (simple for loop reading in four 8-bit values on a 32-bit bus with a 20 MHz processor programmed in C).
The specification for the Transputer boards can be found here.
I think I made four boards, populated three and although I sold the prototype, I gave the other boards to the researchers as they were unable to get funding, which would affect the one researcher’s M.Sc.Eng thesis. After Inmos' demise, there was little market for these boards and I had no use for them without being able to run out of Flash.
Theo van Niekerk went on sabbatical leave during the second semester. I presented his Software and Digits courses and was given a temporary head of department post, however, I was running my own company at the time, and only had to come in for lectures. The steady cash flow always helped and the academic environment had many advantages; friendly staff, test equipment, academic pricing, access to research grants.
The Digits class included microprocessors, but equipment packaged for classroom use was expensive. I wire wrapped a 8051 board on a small prototype card, and then laid out a board for students to use.
More details can be found here. There were two missing links on the power, which were discovered by the first eager student (Warwick Smith, who ten years later would become an extremely productive M.Tech student).
The Software subject used Borland’s Turbo C, with a parallel I/O card in a PC connected to stepper motors, LEDs, switches and analog pots. This must have really made the difference in the students learning, as we covered C, ran several projects and some students even programmed my 8051 boards.
I am not absolutely sure of the date, but I had previously given a course on the 68000. I designed a new course on RISC microprocessors for B.Tech students in evening classes. More details on the MIPS board can be found here FRD PCB layout (the total byte count in the images directory is slightly over a meg).
There is some documentation in trying to buy an Indy workstation in 1995, but by 2002 they were being dumped by other departments as mentioned in the next section (maybe art or physics department at the university next door—not sure as I think the art department used Apple, but there were some SGI boxes there as well).
The Manufacturing Technology Research Centre was attached to the Port Elizabeth Technikon. I was given office space and a “Research Assistant” post by Dr D. Hattingh, with the view of tendering for projects for the Technology Station in Automotive.
Figure 1: A “cast-off” SGI O2 box useful for networking tests. In its day the box was about $15,000 but some research institute pretended to be generous by sending it our way from Pretoria or Johannesburg. Without a compiler it was essentially useless. There was a SGI Indy in a cupboard somewhere else—also discarded by a “generous” department who were not brave enough to tackle the bureaucratic task of writing off an asset and removing the asset tag. My Compaq laptop on the right was running Linux for cross compiling the kernel (ColdFire and MIPS). The PowerPC 823 development was supplied under Cygnus for a Windows platform.
Figure 2: A small embedded network in office. SGI O2 in top right. Network included ColdFire 5272, PowerPC 823, Ampro’s Alchemy Semiconductor Au1500 MIPS on their M3 board, and IDT385 evaluation board (not networked, embedded MIPS testing). My 64-bit PC next to the O2 was running SuSE Linux and used for testing Artisst real-time framework (could not compile for 64-bit so was moved to the laptop). An Ethernet hub is next to the open PC power supply.
Towards the end of 2002 some promising projects arrived—(robotics controller, fruit sorting and classification using digital image processing, intelligent conveyor belt etc.).
The following tasks were performed during this period:-
A full-time post became available in 2003. I remained in the same office. The R300,000 robotics grant was announced at the end of 2003, but took almost a year to come through due to restructuring at government level and bureaucratic haste. I guess funding lead-times are notoriously slow in most parts of the world. Based on the funds announced, the department also granted about R56,000 which went towards the Apple G5, the Avnet Xilinx/AMCC PowerPC GX440 board and a Green Hills Software PowerPC compiler tool chain (at academic prices). The full Xilinx development suite was also purchased at academic pricing. This was a rather busy two years ahead.
Figure 3: (I think this was 2004). My office desk when many toys arrived for the robotics project and department funding. Apple G5 (dual IBM PowerPC 970—64-bit), Celoxica FPGA development platform with Handel-C, AMCC440GX PowerPC with two GbE ports populated but missing the promised Linux port. The Celoxica kit was ordered by Prof. Theo van Niekerk but I had a chance to use it when his Dean duties sucked up his spare time—thanks Theo! The various evaluations are documented elsewhere as the website is being populated. The Celoxica work is not documented as it was fairly way off my research, but I did not see an easy way to port the supporting code to the Virtex-II Pro PowerPC boards that we had. The board was nevertheless very impressive but non-academic pricing of the tools was typical of English companies and they faded soon afterwards.
The following subjects were presented (Automatic Control was for the Mechanical Engineering department, with the first half of the course a PLC component):
|Process Control IV||second||2003||B.Tech|
|Control Systems II||second||2003||Dip|
|Control Systems II||first||2004||Dip|
|Micro Systems Design IV||second||2004||B.Tech|
|Control Systems II||first||2005||Dip|
|Control Systems II||second||2005||Dip|
|Control Systems II||first||2006||Dip|
|Micro Systems Design IV||first||2006||B.Tech|
|Control Systems II||second||2006||Dip|
|Automatic Controls IV||second||2006||Dip|
|Process Control IV||first||2007||B.Tech|
|Automatic Controls IV||first||2007||Dip|
The Control Systems II class was a double class, but each class received the same lectures. Half the subject consisted of “hands on” PLC courses with examinations on the equipment.
I had study leave during the second semester of 2004. During the whole year the “New Lecturer Induction Program and Assessor Training” was held at the Bureau of Education. The changes to assessment were large, which should be interesting to compare the “outcomes based education” in a few years. A detailed report was written up at the end, but the changes would be more deeply felt where the funding relied heavily on government subsidies.
I was a promoter/ exernal examiner together with Andries van der Linder, (M.Dip lecturer who later obtained his PhD in the UK).
The B.Tech project was a six month project that required a write-up or “mini-thesis”. Basil Esterhuysen was also involved with Thapelo Mosebi's project, doing the bulk of the supervision as well as attended the national presentation. Peter was from the Computer Science department; they asked if Electrical Engineering had any joint projects. I was involved in porting Linux to several embedded platforms, so I offered a networking project with the Ethereal (Wire Shark) network analyzer. Peter went to the UK on completing his degree.
The M.Tech projects were generally done by students who had been
working for several years in industry, and either did a project
for their employer who provided funding, or as full-time
students with some financial support from government (basically
paid fees). I was a co-promoter/examiner with Prof. Theo van
Niekerk for Khotso Majara and Basil Esterhuysen, and a
co-promoter/ examiner with Prof. Danie Hattingh for Warwick
Smith. I was Basil Paul's promoter and examiner.
The M.Tech projects were written up as a thesis and generally took two years full-time.
I was involved in various projects at Delta Motor Corporation (General Motors in South Africa during sanctions). I supplied monitoring equipment for strain gauges on large presses and wrote the initial software interface. The graphics software was completed by Dave Forsyth as a contractor.
Herman Vermaak, who was working at Delta, continued the project and registered it as an Masters Diploma, completing his PhD in the Netherlands some years later.
Danie Hattingh wrote a program for replacing numerous tables for a large fan manufacturer. He later obtained a PhD in the UK.
Karl du Preez became Head of Department, Mechanical Engineering at the Technikon.
Before the S7-315 PLCs were delivered, courses were presented on
the S7-200 family. There were ten of each. The S7-315 racks
included ET-2000 I/O on Profibus, and motor on a Profibus
connected variable speed drive and a servo controller. The
HMI was a Siemens graphics display. Several switches and
lamps were connected to I/O cards and the ET-2000.
Besides the Control Systems II, Process Control IV and Automatic Control IV courses, external companies could send staff to complete the Siemens accredited SERV1, SERV2, PRO1 and PRO2 courses. Two to three five-day courses were offered each month during academic holidays (almost half the year).
Non-certified courses were given for the S7-200 Siemens PLCs to introduce students to PLCs. NMMU attendance certificates were issued for courses without any tests, otherwise students had to complete an automation task at the end of the course.
© 2014 — Second Valley Software Pty Ltd. All Rights Reserved